Event semantics and discourse connected-movement in Eastern Cham - - PowerPoint PPT Presentation

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Event semantics and discourse connected-movement in Eastern Cham

Kenneth Baclawski Jr. University of California, Berkeley AFLA 27 August 20, 2020

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 1 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction

At AFLA 26, I proposed a new Ā-feature, discourse connectedness (DC), which refmects a relation between the phrase it attaches to, the sentence it is in, and a prior sentence in the discourse

(Baclawski Jr 2020; Baclawski Jr 2019)

(1)

• a. tʰuːŋ͡m312

Thuận(VN) t̥ɔʔ PROG ŋăʔ make ʔiŋ ʔɔŋ frog năn that ‘Thuận is cooking that frog.’

• b. ʔiŋ ʔɔŋ

frog nănDC that ɲu 3.ANIM ɨŋ want ɓăŋ eat ʔiŋ ʔɔŋ năn ‘That frog, he wants to eat.’1 DC-moved phrases in Eastern Cham must be mentioned in a prior sentence that the current one is interpreted as explaining or elaborating upon

1Words with Chao tone numbers like tʰuːŋ͡m312 represent code-switching or names

in Vietnamese. The third-person animate pronoun ɲu is unspecifjed for number.

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 2 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction: What is discourse subordination?

Figure 1: Discourse structure for (2)

S1 S2 S3

• Disc. coordination
• Disc. subordination
• Disc. subordination

(2)

• a. Kenny is giving a talk on discourse connectedness.
• b. First, he introduced the concept.
• c. Then, he gave a background on Eastern Cham.

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 3 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction: What is discourse subordination?

Figure 1: Discourse structure for (2)

S1 S2 S3

• Disc. coordination
• Disc. subordination
• Disc. subordination

In theories of discourse structure, rhetorical relations between sentences can be subordinating or coordinating

(Grosz & Sidner 1986; Fabricius-Hansen & Ramm 2008 for an overview) Discourse subordination involves expanding upon prior sentences E.g. explanation (answers to Why?), elaboration (What about X) Discourse coordination involves moving onto something new E.g. narration (answers to Then what?), …

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 4 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction: What is discourse connectedness?

(1)

• a. tʰuŋ͡m312

Thuận t̥ɔʔ PROG ŋăʔ make ʔiŋ ʔɔŋ frog năn that ‘Thuận is cooking that frog.’

• b. ʔiŋ ʔɔŋ

frog nănDC that ɲu 3.ANIM ɨŋ want ɓăŋ eat ʔiŋ ʔɔŋ năn ‘That frog, he wants to eat.’ (a ⇓ b) In Eastern Cham, movement of ʔiŋ ʔɔŋ năn ‘that frog’ requires that it be mentioned in a sentence that the current one is discourse subordinate to (i.e. explains or elaborates) Throughout, (a ⇓ b) will be used to mean ‘sentence (b) is interpreted as discourse subordinate to sentence (a)’

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 5 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction: What is discourse connectedness?

(3)

• a. tʰuŋ͡m312

Thuận t̥ɔʔ PROG ŋăʔ make ʔiŋ ʔɔŋ frog năn that ‘Thuận is cooking that frog.’

• b. #jăʔ ni

now ʔiŋ ʔɔŋ frog năn that ɲu 3.ANIM t̥ɔʔ PROG ɓăŋ eat ʔiŋ ʔɔŋ năn INTENDED: ‘Now, that frog, he is eating.’ (a ̸⇓ b) Discourse coordination leads to infelicity When presented with discourses like these, consultants either reject them or insert prior discourse that adds the discourse subordination needed

(e.g. ‘What is Thuận doing in the kitchen?’)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 6 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction: DC as an Ā-feature

I analyzed DC-movement via an Agree relation between a probe

• n C and a feature merged with DPs via (silent) DC-particles, on

analogy with focus particles (Baclawski Jr 2020) [DC] exhibits a range of typical Ā-feature behaviors

Unboundedness Sensitivity to syntactic islands Weak crossover efgects Path containment efgects

(4) [DC DC [ʔiŋ ʔɔŋ frog năn]] that C[uDC] ɲu 3.ANIM ɨŋ want ɓăŋ eat ʔiŋ ʔɔŋ năn ‘That frog, he wants to eat.’ (a ⇓ b) The DC-particle introduces a presupposition that checks whether the phrase it attaches to is mentioned in a prior sentence that the current one is discourse subordinate to (as above)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 7 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Some puzzles

Puzzle #1: What does [DC] mark exactly?

Discourse/rhetorical relations are high-level pragmatic concepts not typically associated with syntactic features ‘Discourse subordination’ is a multifaceted term that encapsulates a constellation of inferences (e.g. Asher & Lascarides 2003)

→ I hypothesize that [DC] marks only one aspect of discourse subordination: a cause or subtype relation between two events in a discourse

Explanations contain an inference that one event caused another (e.g. that the pushing event caused the falling event in (5)) Elaborations contain an inference that one event is a ‘subtype’ of another (as per Asher & Lascarides’ 2003 Segmented Discourse Representation Theory)

(5)

• a. Max fell.
• b. John pushed him. (Asher & Lascarides 2003: 209)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 8 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Some puzzles

Puzzle #2: Why move a DP?

Topic, contrastive topic, and focus are usually framed as statuses of the marked phrase (and its relation with the rest of the sentence) For DC, however, outside of a previous mention requirement, it is the event variable that has the special status In (1b), both ʔiŋ ʔɔŋ ‘frog’ and ɲu ‘3.ANIM’ meet these criteria, and continuations (i.e. 1(c)) can be on the subject of either

(1)

• a. tʰuŋ͡m312

Thuận t̥ɔʔ PROG ŋăʔ make ʔiŋ ʔɔŋ frog năn that ‘Thuận is cooking that frog.’

• b. ʔiŋ ʔɔŋ

frog năn that ɲu 3.ANIM ɨŋ want ɓăŋ eat ʔiŋ ʔɔŋ năn ‘That frog, he wants to eat.’ (a ⇓ b)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 9 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Some puzzles

Puzzle #1: What does [DC] mark exactly? Is Eastern Cham syntax sensitive to the concept of discourse subordination itself? → I hypothesize that [DC] marks only one aspect of discourse subordination: a cause or subtype relation between two events in a discourse Puzzle #2: Why move a DP? → I hypothesize DC-movement is incidentally restricted to DPs in Eastern Cham, but might be expected to involve marking the verb in other languages This talk presents a diagnostic that tests whether DC-movement requires two event variables in the discourse, based on the existential marker hu and the distinction between stage- and individual-level predicates

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 10 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Outline

1

Background on Eastern Cham

2

DC-movement is not contrastive topic

3

DC-movement requires a relation between two events hu as an existential marker Stage- and individual-level predicates Individual-level predicate test for DC-movement

4

Discussion

5

Conclusion

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 11 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Background on Eastern Cham

Eastern Cham is an Austronesian language spoken in south-central Vietnam by about 120,000 people, among whom there is quasi-universal bilingualism with Vietnamese Data were collected via sentence and mini-discourse elicitation from 2014 to 2019 with six main consultants, all native speakers

• f Eastern Cham born and raised in the villages near Phan Rang,

Vietnam

Data for hu as an existential marker and stage- and individual-level predicates were collected from six consultants: three younger consultants with some college education in Ho Chi Minh City and three older consultants with little schooling The individual-level predicate test for DC-movement was conducted with only one, the author’s most experienced consultant, but on multiple difgerent occasions

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 12 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Background on Eastern Cham

Eastern Cham is a largely morphologically isolating SVO language with a tone or register system and many typological characteristics typical of Mainland Southeast Asian languages Transcriptions are in IPA with the exception of an open circle that indicates falling tone/register on the following vowels, in line with the Chamic linguistic tradition (e.g. p̥lɛ̆j ‘buy’ ~ [plɛ̤̆̀j]) Sociolinguistic variation in Eastern Cham and Vietnamese is abstracted in these examples to the most common forms in use in colloquial speech for each word, and the reported local dialect of Vietnamese (Hoàng 1989)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 13 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 2: Map of the Champa Kingdom (Po Dharma 1999, translated)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 14 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DC-movement is not contrastive topicalization

DC-movement may appear similar to contrastive topicalization in languages like English (6)

• a. Who brought what to the potluck?
• b. THE SALADCT… ANTONIOFoc brought the salad.
• c. … And PERSEPHONEFoc brought THE GAZPACHOCT.

(in the spirit of Constant 2014) However, in Eastern Cham:

There is no obvious prosodic contour with DC-movement (Baclawski Jr 2019: 45) Typically continued by more information about the sentence, not a contrasting sentence Not typically used to mark partial answers to QUDs (cf. Büring 2003; Constant 2014)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 15 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contrastive topic in Eastern Cham is marked by hu

More similar to contrastive topicalization is marking by the existential hu Baclawski Jr (2018) argues that subject CTs can be preceded by hu, while hu precedes the verb if there is a CT argument in the predicate (see also Baclawski Jr 2019: 183) (7)

• a. Who invited who to come here?
• b. hu

∃ tʰuːŋ͡m312

CT

Thuận ʔ̥a invite kăwFoc 1SG maj come păʔ ni… here B: ‘THUẬNCT invited MEFoc to come here…’ #B: ‘THUẬNFoc invited MECT to come here…’ b′. tʰuːŋ͡m312

Foc

Thuận hu ∃ ʔ̥a invite kăwCT 1SG maj come păʔ ni… here #B: ‘THUẬNFoc invited MECT to come here…’ B: ‘THUẬNFoc invited MECT to come here…’

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 16 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DC-movement is orthogonal to CT-marking

DC-movement can occur in CT-marking sentences with hu, but

• nly if the corresponding wh-phrase is also DC-moved

In this case, the entire question-answer pair functions as an elaboration or explanation – discourse subordination is necessary for DC-movement. Contrastive topicality is not (8)

• a. jaŋ

which hlɛ̆j CLF.PERSON jŭt friend ʔ̥a invite jaŋ hlɛ̆j maj come păʔ ni here A: ‘Which person did you [friends] invite to come here?’ [Directed at group]

• b. tʰuːŋ͡m312

CT

Thuận kăwFoc 1SG hu ∃ ʔ̥a invite tʰuːŋ͡m312 maj come păʔ ni… here B: ‘ThuậnCT, IFoc invited to come here…’ → DC-movement in Eastern Cham is not contrastive topicalization

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 17 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DC-movement requires a relation between two events

In the introduction, I hypothesized that DC-movement requires a specifjc relation between two events in a discourse This talk will test if that hypothesis is true, based on a test that controls whether relevant event variables are actually introduced in the relevant sentences The next sections give background for and then apply that test

• 1. hu as a general existential marker
• 2. Stage- and individual-level predicates
• 3. The individual-level predicate test for DC-movement

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 18 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu as an existential marker

The Eastern Cham form hu has a variety of uses, all linked to existential semantics, one of which involves existentially closing event variables

(see Thurgood & Li 2007 on hu from the lens of grammaticalization)

These uses closely mirror those of the form adi in Bura (Central Chadic: Nigeria), which Zimmermann (2007) analyzes as a general existential marker

• 1. Thetic sentences that introduce new referents

(9)

• a. kakadu

book adi ∃ ata

• n

kira top tebir table ‘There is a book on the table.’ BURA (Zimmermann 2007: (6b))

• b. hu

∃ tʰa

• ne
• ŋ

grandfather kʰa

• ld

hɔ̆ŋ͡m with mɨ 5 c̥ɛ̆j CLF.ANIMAL cĭm bird ‘There is an old man with fjve birds.’ EASTERN CHAM

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 19 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu as an existential marker

• 2. Existential clefts2

(10)

• a. mda

person adi ∃ ti REL tsa 3SG kuga invite ‘There is somebody that he invited.’ BURA (Zimmermann 2007: (7))

• b. hu

∃ tʰa

• ne

jaŋ person p̥o REL ɲu 3SG ʔ̥a invite ‘There is somebody that he invited.’ EASTERN CHAM

2Zimmermann (2007) reports that Bura clefts with adi can also be identifjcational,

though this does not appear to be the case for Eastern Cham.

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 20 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu as an existential marker

• 3. Emphasis – when used in non-negative, non-cleft sentences, adi

and hu put the existence of an event or situation at issue (11)

• a. thla

cattle adi ∃ akwa in di place nda DEM ya Q ‘IS there cattle in that village?’ BURA (Zimmermann 2007: (43))

• b. hɨ

2SG hu ∃ ʔɨ̆ŋ want ɓăŋ eat ʔɔ̆ʔ mango lɛ̆j Y/N.Q ‘Do you want to eat mango?’ EASTERN CHAM

• c. kăw

1SG hu ∃ ʔɨ̆ŋ want ɓăŋ eat ʔɔ̆ʔ mango ‘I DO want to eat mango.’

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 21 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu as an existential marker

• 4. Negation in many, but not all negative clauses – this will be

explored further in the next section (12)

• a. Pindar

Pindar adi ∃ ata FUT sa drink mbal beer wa NEG ‘Pindar will not drink beer.’ BURA (Zimmermann 2007: (5))

• b. kăw

1SG hu ∃ t̥ɔ̆ʔ PROG ɓăŋ eat lɔ meat nŭʔ chicken ʔo NEG ‘I am not eating chicken.’ EASTERN CHAM

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 22 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu as an existential marker: Summary

Given these parallels, I follow Zimmermann (2007) in analyzing Eastern Cham hu as a general existential marker

• 1. Thetic sentences that introduce new references
• 2. Existential clefts
• 3. Emphasis – when used in non-negative, non-cleft sentences, adi

and hu put the existence of an event or situation at issue

• 4. Negation in many, but not all negative clauses

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 23 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Stage- and individual-level predicates

Carlson (1977) distinguishes between two types of predicates:

Temporary properties of the individuals involved (stage-level predicates) Properties permanent throughout the existence of the individuals involved (individual-level predicates)

This distinction has been argued to be manifested in a wide array

• f constructions cross-linguistically, such as there-insertion:

(13)

• a. There are fjremen available.

(Stage-level predicate)

• b. *There are fjremen altruistic.

(Individual-level predicate)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 24 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Individual-level predicates do not introduce event variables

Kratzer (1995) analyzes the distinction between stage- and individual-level predicates in terms of argument structure:

Stage-level predicates have an argument position for Davidsonian event variables Individual-level predicates do not

A variety of linguistic manifestations of the stage- and individual-level predicate distinction can be accounted for by the argument structure requirements of the construction in question (Kratzer 1995; Fe00)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 25 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu+negation as a diagnostic for event variables in Eastern Cham

Turning back to hu, both hu and Bura adi must appear in negative clauses with stage-level predicates Zimmermann (2007) argues that negation, at least in Bura, requires that all variables be existentially closed, including the

• pen event variable

(see Zeijlstra 2004 and others on existential closure and negation)

hu/adi is the only way to supply that existential closure (14)

• a. Pindar

Pindar adi ∃ ata FUT e sa drink mbal beer wa NEG ‘Pindar will not drink beer.’ BURA (Zimmermann 2007: (5))

• b. kăw

1SG hu ∃ t̥ɔ̆ʔ PROG e ɓăŋ eat lɔ meat nŭʔ chicken ʔo NEG ‘I am not eating chicken.’ EASTERN CHAM

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 26 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu+negation as a diagnostic for event variables in Eastern Cham

Negative clauses with individual-level predicates cannot have hu/adi, as they do not introduce event variables that need existential closure under negation Both in Bura and Eastern Cham, the verb ‘know’ is found in such individual-level predicates (15)

• a. Musa

Musa asinda know mya language Bura Bura wa NEG ‘Musa does not understand / cannot speak Bura.’ BURA (Zimmermann 2007: (25))

• b. kăw

1SG thăw know ʔo NEG ‘I don’t know.’ EASTERN CHAM

• c. kăw

1SG ɲŭm drink ka21 fe33 cofgee c̥ɨŋ ABLE ʔo NEG ‘I can’t drink cofgee.’

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 27 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu+negation as a diagnostic for event variables in Eastern Cham

In the absence of hu, negation is either marked solely by ʔo or by an additional ʄăwʔ ‘correct’ (16)

• a. mɛ̆j

female năn that hu ∃ p̥uj p̥ɛ be.happy ʔo NEG ‘That woman is not happy.’ (Stage-level)

• b. mɛ̆j

female năn that ʄăwʔ correct tʰa

• ne

jaŋ CLF.PERSON p̥uj p̥ɛ happy ʔo NEG ‘That woman is not a happy person.’ (Individual-level)

• c. kăw

1SG hu ∃ t̥ɔʔ live păʔ in mlɛ̆j village năn that ʔo NEG ‘I do not live in that village [temporarily].’ (Stage-level)

• d. kăw

1SG (ʄăwʔ) correct nɨ̆ʔ be.born păʔ in mlɛ̆j village năn that ʔo NEG ‘I was not born in that village.’ (Individual-level)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 28 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

hu+negation as a diagnostic for event variables in Eastern Cham

Jäger (2001) argues that the distinction between stage- and individual-level predicates masks a variety of difgerent contrasts, including largely pragmatic ones I will focus on the interaction between hu and negation as a language-internal diagnostic for the existence of event variables in Eastern Cham (see Baclawski Jr 2019: 31 for similar argumentation)

Consultants provide clear judgments between hu, ∅, and ʄăwʔ Kratzer’s (1995) account gives a clear explanation for why a general existential marker should be involved in some negative clauses, but not others

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 29 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Individual-level predicate test for DC-movement

Putting the preceding sections together: if DC-movement requires a specifjc relation between two events in a discourse, a prior event e1 and a current event e2:

DC-movement should be impossible in a sentence consisting only

• f an individual-level predicate (there is no e2!)

DC-movement should be impossible if the relevant prior sentence consists only of an individual-level predicate (there is no e1!)

(17) PREDICTIONS:

• 1. ✓DC-phrase stage-level predicate DC-phrase
• 2. ✗DC-phrase individual-level predicate DC-phrase
• 3. Prior sentence: stage-level predicate

✓DC-phrase stage-level predicate DC-phrase

• 4. Prior sentence: individual-level predicate

✗DC-phrase stage-level predicate DC-phrase

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 30 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Individual-level predicate test: Predictions #1–2

Predictions #1–2: DC-marking should be possible in sentences with stage-level predicates, but not in those with only individual-level predicates This prediction is borne out:3 (18)

• a. mlɛ̆j

village nănDC that kăw 1SG t̥ɔʔ live păʔ in mlɛ̆j năn ‘That village, I live in [temporarily].’ (Stage-level)

• b. *mlɛ̆j

village nănDC that kăw 1SG nɨ̆ʔ be.born păʔ in mlɛ̆j năn INTENDED: ‘That village, I was born in.’ (Individual-level)

3Note that the preposition păʔ ‘in’ is dropped when argument prepositional

phrases are Ā-moved in Eastern Cham (cf. Wang 2007 on p-drop in Indonesian).

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 31 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Individual-level predicate test: Predictions #1–2

As a control, both sentences above can be interpreted as explaining or elaborating upon prior sentences in Eastern Cham, if explicitly marked by jwa ‘because’ (without DC-movement) → Discourse subordination itself does not require that two event variables be introduced in the discourse! (19)

• a. kăw

1SG t̥ɔʔ live păʔ in mlɛ̆j village năn that t̥om many tʰŭn year hl̥aw before ‘I have lived in this village for many years.’

• b. jwa

because kăw 1SG nɨ̆ʔ be.born păʔ in mlɛ̆j village năn that ‘Because I was born in that village.’ (Individual-level)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 32 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Individual-level predicate test: Predictions #3–4

DC-marking should also not be possible if the prior sentence contains an individual-level predicate This prediction is also borne out: (20)

• a. kăw

1SG t̥ɔʔ live păʔ in p̥aj k̥ɔl Saigon ‘I live in Saigon [temporarily].’ (Stage-level)

• b. p̥aj k̥ɔlDC

Saigon mɛʔ mɨ parent kăw 1SG naw go ŋăʔ make p̥jŭʔ work păʔ in p̥aj k̥ɔl ‘My parents went to work in Saigon.’ (21)

• a. kăw

1SG nɨ̆ʔ be.born păʔ in p̥aj k̥ɔl Saigon ‘I was born in Saigon.’ (Individual-level)

• b. #p̥aj k̥ɔlDC

Saigon mɛʔ mɨ parent kăw 1SG naw go ŋăʔ make p̥jŭʔ work păʔ in p̥aj k̥ɔl INTENDED: ‘My parents went to work in Saigon.’

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 33 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Individual-level predicate test: Predictions #3–4

Again, as a control, both (a) sentences can be continued by the (b) sentence if explicitly marked by jwa ‘because’ and no DC-movement (22)

• a. kăw

1SG t̥ɔʔ live păʔ in p̥aj k̥ɔl Saigon ‘I live in Saigon [temporarily].’ (Stage-level) a′. kăw 1SG nɨ̆ʔ be.born păʔ in p̥aj k̥ɔl Saigon ‘I was born in Saigon.’ (Individual-level)

• b. jwa

because mɛʔ mɨ parent kăw 1SG naw go ŋăʔ make p̥jŭʔ work păʔ in p̥aj k̥ɔl Saigon ‘Because my parents went to work in Saigon.’

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 34 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Discussion

DC-movement fails in the absence of two semantic event variables in the relevant prior and current sentences

[DC] marks a specifjc relation between two events (‘cause’ via explanation or ‘subtype’ via elaboration) [DC] does not purely mark discourse subordination, as the sentences above can all be interpreted as explanations (with jwa ‘because’ and without DC-movement)

→ Syntax need not be sensitive to broad types of discourse/rhetorical relations, but it should be sensitive to relations between events in a discourse (Puzzle #1) → We need a dynamic model of event semantics that allows syntax to see prior events in the discourse, their participants, and the relations between those events (perhaps limited to broad categories along the lines of discourse subordination and coordination)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 35 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Discussion

A new puzzle: why ‘cause’ and ‘subtype’? In the philosophy of explanation, two notions are of relevance:

‘Causal explanation’, which corresponds with explanation (i.e. answers to 23a) ‘Constitutive explanation’, which corresponds with elaboration (i.e. answers to 23b) Why is the glass fragile? is ambiguous between the two readings

(23)

• a. How did the glass become fragile?
• b. What makes the glass fragile? (Ylikoski 2013: 279)

Ylikoski (2013) demonstrates that these two types of explanation have difgerent properties, but share core ideas about explanation Therefore, [DC] tracks the presence of one type of relation between events (those associated with explanations)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 36 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Discussion

→ DC-moved phrases do not have a special status – it is the events of which they are participants that has a special status (Puzzle #2) Why does DC-movement not involve verbal marking or VP-movement?

As for the former, there is no bound morphology to recruit (Catalan clitic right-dislocation, which exhibits very similar pragmatic properties to DC-movement in Eastern Cham, involves verbal clitics; see Baclawski Jr 2019: 167) As for the latter, there is VP-fronting in Eastern Cham, but more research is needed into its interpretation. If its interpretation is difgerent from DC, perhaps Eastern Cham recruits DP-movement instead, which does not mark other pragmatics like contrastive topicalization

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 37 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Other possible analyses

The above analysis relies on hu as an existential marker and Kratzer’s (1995) account of individual-level predicates (that they lack event variables) Jäger (2001) proposes that some contrasts associated with the stage- and individual-level predicate distinction are due to the topic-comment structure of difgerent predicates

For stage-level predicates, subjects are optionally topics For individual-level predicates, subjects are necessarily topics The restriction of individual-level predicates then accounts for its ungrammaticality or limited readings in difgerent constructions

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 38 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Other possible analyses

But, topic-comment structure does not appear relevant for DC-movement Contrastive topic is orthogonal to DC-movement (Section 2) Additionally, in (1b), the subject is interpretable as an old information and/or aboutness topic (1)

• a. tʰuːŋ͡m312

Thuận(VN) t̥ɔʔ PROG e1 ŋăʔ make ʔiŋ ʔɔŋ frog năn that ‘Thuận is cooking that frog.’

• b. ʔiŋ ʔɔŋ

frog nănDC that ɲu 3.ANIM ɨŋ want e2 ɓăŋ eat ʔiŋ ʔɔŋ năn ‘That frog, he wants to eat.’ (a ⇓ b)

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 39 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Other possible analyses

Multiple phrases can also be DC-moved from the same verb phrase (24) (24) han cake ni this nɨ̆ʔ child mɛ̆j female sĭt small năn that tʰuːŋ͡m312 Thuận ʔ̥a invite nɨ̆ʔ mɛ̆j sĭt năn maj come ɓăŋ eat han ni ‘This cake, Thuận invited that little girl to come eat.’ Therefore, the status of one phrase in a sentence as a topic, contrastive topic, or DC-phrase does not appear to impede DC-movement of another phrase Additionally, any alternative analysis would have to provide an account of why the existential marker hu appears in some, but not all negative clauses in such a way that aligns with DC-movement

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 40 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Conclusion

Syntactic movement can be sensitive to relations between events introduced in a discourse (specifjcally, explanative relations) As a corollary, we need a dynamic model of events in a discourse, their respective participants, and their respective event relations

Perhaps not dissimilar from the directed graph from the introduction

Conversely, syntactic movement does not need to be sensitive to broad discourse/rhetorical relation categories like discourse subordination, at least in this case Pragmatic syntactic movement of phrases need not indicate a special status of the moved phrase – Eastern Cham DC-movement indicates a special status of the event the moved phrase is a participant of

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 41 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Thank you!

My sincere thanks to the Cham people of Phan Rang, Vietnam, especially to Sakaya (Professor Trường Văn Món) and Sikhara (Hamu Ligaih), without whom this work would not be possible. All the many mistakes are my own Thanks also to Peter Jenks, Line Mikkelsen, and Seth Yalcin for their extensive help and feedback, in addition to audiences at AFLA 26 and UC Berkeley

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 42 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References I

Asher, Nicholas & Alex Lascarides. 2003. Logics of conversation. Cambridge University Press. Baclawski Jr, Kenneth. 2018. Contrastive topic in Eastern Cham. Berkeley Formal Papers in Linguistics 1(1). https://escholarship.org/uc/item/3hp0s863. Baclawski Jr, Kenneth. 2019. Discourse connectedness: The syntax–discourse structure interface. University of California, Berkeley dissertation. Baclawski Jr, Kenneth. 2020. Optional wh-movement and topicalization in Eastern Cham. In Ileana Paul (ed.), Proceedings of the Twenty-Sixth Meeting

• f the Austronesian Formal Linguistics Association (AFLA). University of

Western Ontario, 1–17. Büring, Daniel. 2003. On d-trees, beans, and b-accents. Linguistics and Philosophy 26. 511–546. Carlson, Greg N. 1977. Reference to kinds in English. University of Massachusetts dissertation. Constant, Noah. 2014. Contrastive topic: Meanings and realizations. University

• f Massachusetts, Amherst dissertation.

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 43 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References II

Fabricius-Hansen, Cathrine & Wiebke Ramm. 2008. Editor’s introduction: Subordination and coordination from difgerent perspectives. Cathrine Fabricius-Hansen & Wiebke Ramm (eds.). J. Benjamins. 1–30. https://doi.org/10.1075/slcs.98.01fab. Grosz, Barbara & Candace Sidner. 1986. Attention, intentions, and the structure of discourse. Computational Linguistics 12(3). 174–204. Hoàng, Thị Châu. 1989. Tiếng Việt trên các Miền Đát Nước [The Vietnamese Language in the Regions of the Country]. Hà Nội: Nhà Xuất Bản Khoa Học Xã Hội. Jäger, Gerhard. 2001. Topic‐comment structure and the contrast between stage level and individual level predicates. Journal of Semantics 18(2). 83–126. https://doi.org/10.1093/jos/18.2.83. Kratzer, Angelika. 1995. Stage-level and individual-level predicates. In Gregory N. Carlson & Francis Jefgry Pelletier (eds.), The Generic Book, 125–175. University of Chicago Press.

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 44 / 45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References III

Thurgood, Graham & Fengxiang Li. 2007. The grammaticalization paths of Proto-Chamic *hu “receive”. In Proceedings of the Chicago Linguistics Society 39, 205–214. Wang, Chyan-an Arthur. 2007. Sluicing and resumption. In Proceedings of the 37th North East Linguistic Society, 239–252. Amherst, MA: GLSA. Ylikoski, Petri. 2013. Causal and constitutive explanation compared. Erkenn

• 78. 277–297. https://doi.org/10.1007/s10670-013-9513-9.

Zeijlstra, Hedde. 2004. Sentential negation and negative concord. Netherlands Graduate School of Linguistics. Zimmermann, Malte. 2007. Overt existential closure in Bura (Central Chadic). In T. Friedman & M. Gibson (eds.), Proceedings of SALT 17, 333–350. Cornell, CLC Publications.

Events and DC-movement in Eastern Cham AFLA 27 August 20, 2020 45 / 45